app.py

import gradio as gr
from huggingface_hub import InferenceClient

client = InferenceClient("thviet79/model-QA-medical")

def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    messages = [{"role": "system", "content": system_message}]

    for val in history:
        if val[0]:
            messages.append({"role": "user", "content": val[0]})
        if val[1]:
            messages.append({"role": "assistant", "content": val[1]})

    messages.append({"role": "user", "content": message})

    # Here, call the Hugging Face Inference API with the messages and other parameters
    response = client.text_generation(
        model="thviet79/model-QA-medical",
        inputs=messages,
        parameters={"max_tokens": max_tokens, "temperature": temperature, "top_p": top_p},
    )
    return response["generated_text"]  # Extract the model's response

demo = gr.ChatInterface(
    respond,
    additional_inputs=[
        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
        gr.Slider(
            minimum=0.1,
            maximum=1.0,
            value=0.95,
            step=0.05,
            label="Top-p (nucleus sampling)",
        ),
    ],
    type='messages'  # This makes it use the OpenAI-style structure
)

if __name__ == "__main__":
    demo.launch(share = True)